File size: 13,333 Bytes
7db0ae4 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 |
import requests, types, time
import json, uuid
import traceback
from typing import Optional
import litellm
import httpx, aiohttp, asyncio
from .prompt_templates.factory import prompt_factory, custom_prompt
class OllamaError(Exception):
def __init__(self, status_code, message):
self.status_code = status_code
self.message = message
self.request = httpx.Request(method="POST", url="http://localhost:11434")
self.response = httpx.Response(status_code=status_code, request=self.request)
super().__init__(
self.message
) # Call the base class constructor with the parameters it needs
class OllamaConfig:
"""
Reference: https://github.com/jmorganca/ollama/blob/main/docs/api.md#parameters
The class `OllamaConfig` provides the configuration for the Ollama's API interface. Below are the parameters:
- `mirostat` (int): Enable Mirostat sampling for controlling perplexity. Default is 0, 0 = disabled, 1 = Mirostat, 2 = Mirostat 2.0. Example usage: mirostat 0
- `mirostat_eta` (float): Influences how quickly the algorithm responds to feedback from the generated text. A lower learning rate will result in slower adjustments, while a higher learning rate will make the algorithm more responsive. Default: 0.1. Example usage: mirostat_eta 0.1
- `mirostat_tau` (float): Controls the balance between coherence and diversity of the output. A lower value will result in more focused and coherent text. Default: 5.0. Example usage: mirostat_tau 5.0
- `num_ctx` (int): Sets the size of the context window used to generate the next token. Default: 2048. Example usage: num_ctx 4096
- `num_gqa` (int): The number of GQA groups in the transformer layer. Required for some models, for example it is 8 for llama2:70b. Example usage: num_gqa 1
- `num_gpu` (int): The number of layers to send to the GPU(s). On macOS it defaults to 1 to enable metal support, 0 to disable. Example usage: num_gpu 0
- `num_thread` (int): Sets the number of threads to use during computation. By default, Ollama will detect this for optimal performance. It is recommended to set this value to the number of physical CPU cores your system has (as opposed to the logical number of cores). Example usage: num_thread 8
- `repeat_last_n` (int): Sets how far back for the model to look back to prevent repetition. Default: 64, 0 = disabled, -1 = num_ctx. Example usage: repeat_last_n 64
- `repeat_penalty` (float): Sets how strongly to penalize repetitions. A higher value (e.g., 1.5) will penalize repetitions more strongly, while a lower value (e.g., 0.9) will be more lenient. Default: 1.1. Example usage: repeat_penalty 1.1
- `temperature` (float): The temperature of the model. Increasing the temperature will make the model answer more creatively. Default: 0.8. Example usage: temperature 0.7
- `stop` (string[]): Sets the stop sequences to use. Example usage: stop "AI assistant:"
- `tfs_z` (float): Tail free sampling is used to reduce the impact of less probable tokens from the output. A higher value (e.g., 2.0) will reduce the impact more, while a value of 1.0 disables this setting. Default: 1. Example usage: tfs_z 1
- `num_predict` (int): Maximum number of tokens to predict when generating text. Default: 128, -1 = infinite generation, -2 = fill context. Example usage: num_predict 42
- `top_k` (int): Reduces the probability of generating nonsense. A higher value (e.g. 100) will give more diverse answers, while a lower value (e.g. 10) will be more conservative. Default: 40. Example usage: top_k 40
- `top_p` (float): Works together with top-k. A higher value (e.g., 0.95) will lead to more diverse text, while a lower value (e.g., 0.5) will generate more focused and conservative text. Default: 0.9. Example usage: top_p 0.9
- `system` (string): system prompt for model (overrides what is defined in the Modelfile)
- `template` (string): the full prompt or prompt template (overrides what is defined in the Modelfile)
"""
mirostat: Optional[int] = None
mirostat_eta: Optional[float] = None
mirostat_tau: Optional[float] = None
num_ctx: Optional[int] = None
num_gqa: Optional[int] = None
num_thread: Optional[int] = None
repeat_last_n: Optional[int] = None
repeat_penalty: Optional[float] = None
temperature: Optional[float] = None
stop: Optional[
list
] = None # stop is a list based on this - https://github.com/jmorganca/ollama/pull/442
tfs_z: Optional[float] = None
num_predict: Optional[int] = None
top_k: Optional[int] = None
top_p: Optional[float] = None
system: Optional[str] = None
template: Optional[str] = None
def __init__(
self,
mirostat: Optional[int] = None,
mirostat_eta: Optional[float] = None,
mirostat_tau: Optional[float] = None,
num_ctx: Optional[int] = None,
num_gqa: Optional[int] = None,
num_thread: Optional[int] = None,
repeat_last_n: Optional[int] = None,
repeat_penalty: Optional[float] = None,
temperature: Optional[float] = None,
stop: Optional[list] = None,
tfs_z: Optional[float] = None,
num_predict: Optional[int] = None,
top_k: Optional[int] = None,
top_p: Optional[float] = None,
system: Optional[str] = None,
template: Optional[str] = None,
) -> None:
locals_ = locals()
for key, value in locals_.items():
if key != "self" and value is not None:
setattr(self.__class__, key, value)
@classmethod
def get_config(cls):
return {
k: v
for k, v in cls.__dict__.items()
if not k.startswith("__")
and not isinstance(
v,
(
types.FunctionType,
types.BuiltinFunctionType,
classmethod,
staticmethod,
),
)
and v is not None
}
# ollama implementation
def get_ollama_response(
api_base="http://localhost:11434",
model="llama2",
messages=None,
optional_params=None,
logging_obj=None,
acompletion: bool = False,
model_response=None,
encoding=None,
):
if api_base.endswith("/api/chat"):
url = api_base
else:
url = f"{api_base}/api/chat"
## Load Config
config = litellm.OllamaConfig.get_config()
for k, v in config.items():
if (
k not in optional_params
): # completion(top_k=3) > cohere_config(top_k=3) <- allows for dynamic variables to be passed in
optional_params[k] = v
optional_params["stream"] = optional_params.get("stream", False)
data = {"model": model, "messages": messages, **optional_params}
## LOGGING
logging_obj.pre_call(
input=None,
api_key=None,
additional_args={
"api_base": url,
"complete_input_dict": data,
"headers": {},
"acompletion": acompletion,
},
)
if acompletion is True:
if optional_params.get("stream", False) == True:
response = ollama_async_streaming(
url=url,
data=data,
model_response=model_response,
encoding=encoding,
logging_obj=logging_obj,
)
else:
response = ollama_acompletion(
url=url,
data=data,
model_response=model_response,
encoding=encoding,
logging_obj=logging_obj,
)
return response
elif optional_params.get("stream", False) == True:
return ollama_completion_stream(url=url, data=data, logging_obj=logging_obj)
response = requests.post(
url=f"{url}",
json=data,
)
if response.status_code != 200:
raise OllamaError(status_code=response.status_code, message=response.text)
## LOGGING
logging_obj.post_call(
input=messages,
api_key="",
original_response=response.text,
additional_args={
"headers": None,
"api_base": api_base,
},
)
response_json = response.json()
## RESPONSE OBJECT
model_response["choices"][0]["finish_reason"] = "stop"
if data.get("format", "") == "json":
message = litellm.Message(
content=None,
tool_calls=[
{
"id": f"call_{str(uuid.uuid4())}",
"function": {
"arguments": response_json["message"]["content"],
"name": "",
},
"type": "function",
}
],
)
model_response["choices"][0]["message"] = message
else:
model_response["choices"][0]["message"] = response_json["message"]
model_response["created"] = int(time.time())
model_response["model"] = "ollama/" + model
prompt_tokens = response_json["prompt_eval_count"] # type: ignore
completion_tokens = response_json["eval_count"]
model_response["usage"] = litellm.Usage(
prompt_tokens=prompt_tokens,
completion_tokens=completion_tokens,
total_tokens=prompt_tokens + completion_tokens,
)
return model_response
def ollama_completion_stream(url, data, logging_obj):
with httpx.stream(
url=url, json=data, method="POST", timeout=litellm.request_timeout
) as response:
try:
if response.status_code != 200:
raise OllamaError(
status_code=response.status_code, message=response.iter_lines()
)
streamwrapper = litellm.CustomStreamWrapper(
completion_stream=response.iter_lines(),
model=data["model"],
custom_llm_provider="ollama_chat",
logging_obj=logging_obj,
)
for transformed_chunk in streamwrapper:
yield transformed_chunk
except Exception as e:
raise e
async def ollama_async_streaming(url, data, model_response, encoding, logging_obj):
try:
client = httpx.AsyncClient()
async with client.stream(
url=f"{url}", json=data, method="POST", timeout=litellm.request_timeout
) as response:
if response.status_code != 200:
raise OllamaError(
status_code=response.status_code, message=response.text
)
streamwrapper = litellm.CustomStreamWrapper(
completion_stream=response.aiter_lines(),
model=data["model"],
custom_llm_provider="ollama_chat",
logging_obj=logging_obj,
)
async for transformed_chunk in streamwrapper:
yield transformed_chunk
except Exception as e:
traceback.print_exc()
async def ollama_acompletion(url, data, model_response, encoding, logging_obj):
data["stream"] = False
try:
timeout = aiohttp.ClientTimeout(total=litellm.request_timeout) # 10 minutes
async with aiohttp.ClientSession(timeout=timeout) as session:
resp = await session.post(url, json=data)
if resp.status != 200:
text = await resp.text()
raise OllamaError(status_code=resp.status, message=text)
response_json = await resp.json()
## LOGGING
logging_obj.post_call(
input=data,
api_key="",
original_response=response_json,
additional_args={
"headers": None,
"api_base": url,
},
)
## RESPONSE OBJECT
model_response["choices"][0]["finish_reason"] = "stop"
if data.get("format", "") == "json":
message = litellm.Message(
content=None,
tool_calls=[
{
"id": f"call_{str(uuid.uuid4())}",
"function": {
"arguments": response_json["message"]["content"],
"name": "",
},
"type": "function",
}
],
)
model_response["choices"][0]["message"] = message
else:
model_response["choices"][0]["message"] = response_json["message"]
model_response["created"] = int(time.time())
model_response["model"] = "ollama/" + data["model"]
prompt_tokens = response_json["prompt_eval_count"] # type: ignore
completion_tokens = response_json["eval_count"]
model_response["usage"] = litellm.Usage(
prompt_tokens=prompt_tokens,
completion_tokens=completion_tokens,
total_tokens=prompt_tokens + completion_tokens,
)
return model_response
except Exception as e:
traceback.print_exc()
raise e
|